[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

From |
Stas Kolenikov <skolenik@gmail.com> |

To |
statalist@hsphsun2.harvard.edu |

Subject |
Re: st: Fama-MacBeth regressions |

Date |
Wed, 13 Oct 2004 14:40:20 -0400 |

On Tue, 12 Oct 2004 15:12:49 -0400, Michael Blasnik <michael.blasnik@verizon.net> wrote: > Well, I can't really comment much on the Fama MacBeth approach since I never > heard of it before today and work in a completely different field. But, I > believe that slopes-as-outcomes or stagewise regression can work quite well > in certain contexts -- particularly when the first stage regressions have > little uncertainty and serve to get at the quantity of interest. In my work > analyzing energy usage of buildings over time, the first stage regression > (for each building for each year) is merely to adjust for weather > differences (heating degree days) between years and the individual OLS > models typically have an R^2 of 0.95. The results from those regressions > are then used (indirectly through predicted outcomes) as the dependent > variable to analyze differences between buildings. I realize that this > analysis may be more efficiently (in a statistical sense) done within a > multilevel modeling framework, but I think little is lost by simplifying it > into stages since the first level models have such good fits while the gains > from the stage-wise approach are substantial in terms of computing time and, > in some ways, interpretability. These advantages are particularly > noticeable when dealing with tens of thousands of buildings at once. I'm > not sure how long -gllamm- might take to fit a multi-level model with 20,000 > cross sections and 500,000 observations...do you think it's worth trying or > will it take several weeks/months/years? Well you can fit one just for fun... my guess would be that it would take about a month for a simple model like a couple of regressors at each level. But with R^2 of 0.95 I myself would not bother much about any corrections. The advantage is that you would get efficient estimates and an idea about the size of effects at each level, but if that is not in the focus of your research, then it may not be worth it. Have a look at Murphy and Topel paper in JBES, they provide analytical corrections for the standard errors. You can try them out, but I would be surprised if you would see an increase in the standard errors by more than 10% -- and that is what you would see when you switch to -robust- option, anyway. -- Stas Kolenikov http://stas.kolenikov.name * * For searches and help try: * http://www.stata.com/support/faqs/res/findit.html * http://www.stata.com/support/statalist/faq * http://www.ats.ucla.edu/stat/stata/

**References**:**st: Fama-MacBeth regressions***From:*"Alex Edmans" <aedmans@mit.edu>

**Re: st: Fama-MacBeth regressions***From:*Stas Kolenikov <skolenik@gmail.com>

**Re: st: Fama-MacBeth regressions***From:*"Subhankar Nayak" <nayak@bellsouth.net>

**Re: st: Fama-MacBeth regressions***From:*Stas Kolenikov <skolenik@gmail.com>

**Re: st: Fama-MacBeth regressions***From:*"Michael Blasnik" <michael.blasnik@verizon.net>

- Prev by Date:
**Re: st: biprobit** - Next by Date:
**Re: st: Text editors FAQ update** - Previous by thread:
**RE: st: Fama-MacBeth regressions** - Next by thread:
**st: RE: translation between binary system and the decimal system.** - Index(es):

© Copyright 1996–2014 StataCorp LP | Terms of use | Privacy | Contact us | What's new | Site index |